Nowadays, when multiple aspects of our life depend on complex cyber-physical systems, automated anomaly detection, prevention and handling is a critical issue that inuence our security and quality of life. Recent catastrophic events showed that manual (human-based) handling of anomalies in complex systems is not recommended, automatic and intelligent handling being the proper approach. This paper presents, through a number of case studies, the challenges and possible solutions for implementing computer-based anomaly detection systems.
Time-series are ordered sequences of discrete-time data. Due to their temporal dimension, anomaly detection techniques used in time-series have to take into consideration time correlations and other time-related particularities. Generally, in order to evaluate the quality of an anomaly detection technique, the confusion matrix and its derived metrics such as precision and recall are used. These metrics, however, do not take this temporal dimension into consideration. In this paper, we propose three metrics that can be used to evaluate the quality of a classification, while accounting for the temporal dimension found in time-series data.
Conservation of fruit tree biodiversity is important for the mankind according to the Convention on Biological Diversity. In Romania, due to favorable environmental conditions, numerous genetic resources of plum, apple, walnut, hazelnut, sweet chestnut and other
fruit crops are present. Identification, evaluation and conservation of fruit genetic resources activities were launched in 1970’ in order to limit the loss of the biodiversity due to erosion and genetic vulnerability. Fruit Growing Research & Extension Station (SCDP) of Valcea, which is belonging now to the University of Craiova, was assigned to deal with conservation of genetic resources for the Prunus, Juglans, Corylus and Castanea genera. As result, national hazelnut collection, the sweet chestnut collection and a part of the plum and walnut national collections are located here. Genetic resources of Malus, Pyrus, Sambucus, Carya and Salix are hold in the active collections too. The ex situ collections include 1160 accessions, out of those 48 species, 533 cultivars and 565 other types like hybrids, biotypes, mutants, etc. Autochthonous or ‘original’ accessions include 173 local cultivars and 565 other genotypes. Breeding activity based on valuable germplasm conducted to releasing of a total number of 31 cultivars and 8 rootstocks registered in the Romanian Official Catalogue for Varieties, 15 varieties have been patented in Romania and for one by CPVO. Identification, in situ evaluation, collection, ex situ evaluation, propagation and regeneration activities regarding fruit tree genetic resources have to be continued in order to conserve the local fruit tree biodiversity and to value it through breeding and use of the new varieties in the orchards.
Technology is expanding at a speed previously unsurpassed; therefore an emphasis is made on integration, optimization and increasing efficiency on different fields. Business intelligence is a new concept which became popular alongside online marketing, as both use external and internal data in order to make better decisions, process improvement and optimizations. The aim of this paper is to propose a solution for business intelligence in the commercial field based on a model of internal sales platform. Gathering data for all the structures integrated in this platform improves the optimization of products and services offered by the companies and also streamline the process of sales. This business solution underpins another solution enabling market research on consumer preferences in order to lead to the development of new products / services according to the real needs of the market at the present time.
In the 2016 revision of the World Health Organization classification the term therapy-related myeloid neoplasia (t-MN) defines a subgroup of acute myeloid leukemia (AML) comprising patients who develop myelodysplastic syndrome (MDS-t) or acute myeloid leukemia (AML-t) after treatment with cytotoxic and/or radiation therapy for various malignancies or autoimmune disorders. We report the case of a 36 year old patient with t-MN (t-MDS) after achieving complete remission (CR) of a PML-RARA positive acute promyelocytic leukemia (APL) at 32 months after diagnosis. Initially classified as low risk APL and treated according to the AIDA protocol - induction and 3 consolidation cycles - the patient achieved a complete molecular response in September 2013 and started maintenance therapy. On follow-up PML-RARA transcript remained negative. In January 2016 leukopenia and thrombocytopenia developed and a peripheral blood smear revealed hypogranular and agranular neutrophils. Immunophenotyping in the bone marrow aspirate identified undifferentiated blast cells that did not express cytoplasmic myeloperoxidase. The cytogenetic study showed normal karyotype. The molecular biology tests not identified PMLRARA transcript. A diagnosis of t-MDS (AREB-2 - WHO 2008) was established. Treatment of AML was started with 2 “3+7” regimens and 1 MEC cycle. Two months from diagnosis, while in CR, an allogeneic HSCT from an unrelated HLA compatible donor was performed after myeloablative regimen. An unfavorable clinical evolution was followed by death on day 9 after transplantation. The occurrence of t-MNs during CR of APL represents a particular problem in terms of follow-up and differential diagnosis of relapse and constitutes a dramatic complication for a disease with a favorable prognosis.
This work was supported by the grants PN 41-087 /PN2-099 from the Romanian Ministry of Research and Technology